AI Autonomous Agri-Drones with Smart Soil and Crop Health Monitoring

Abstract

Agriculture is being rapidly modernized through the inclusion of Artificial Intelligence, Internet of Things, and Unmanned Aerial Vehicles. Current systems, however, do not have integrated autonomous aerial monitoring with ground sensor intelligence for real-time actionable crop and soil health information. In this article, a new system of combining autonomous farming drones equipped with AI technology with smart networks of soil sensors for real-time monitoring, analysis, and action on farm data is proposed. Leveraging multispectral vision, AI- enabled disease identification, real-time soil parameter measurement, and actuation via drone precision, this system resolves crucial scaling, responsiveness, and sustainability constraints of current agriculture. The research describes the architecture of the system, principal technologies, implementation strategy, and gives a comparative review of its benefits compared to available models.

Introduction

The paper presents an AI-driven hybrid system for sustainable agriculture that integrates autonomous drones with AI functionalities and IoT-based soil sensors. This system enables real-time, dynamic, and intelligent decision-making by combining continuous aerial surveillance with precise soil condition monitoring and automated drone actuation (e.g., spraying), addressing limitations of existing isolated or non-scalable systems.

Methodology:

• System Design: Autonomous drones equipped with RGB, multispectral, and thermal cameras capture aerial data, while IoT soil sensors monitor moisture, temperature, pH, and humidity. Data is processed centrally for precision farming decisions.

• Data Acquisition & Preprocessing: Ground sensors and drones collect geo-tagged and timestamped data, which undergoes normalization, noise removal, and augmentation to prepare for machine learning.

• AI Integration: Multiple AI models are used—CNNs for crop health analysis from images, XG-Boost for soil data regression, and LSTMs for temporal predictions.

• Decision Support: The system offers real-time recommendations and autonomous drone actions like watering or pesticide spraying, with options for manual override.

• User Interface: Web and mobile apps provide dashboards, alerts, and interactive features to farmers, supporting multiple languages and voice commands.

• Evaluation: Metrics like accuracy, precision, MAE, RMSE, and system responsiveness assess performance.

Datasets: The system uses multimodal data combining time-series soil sensor data and aerial imagery to train models for precise crop health and resource management.

Results: The gradient boosting model achieved high accuracy in predicting irrigation and fertilization needs, reducing water usage by 35% and fertilizer by 25% in simulations, demonstrating the system’s effectiveness in optimizing resource use and detecting crop diseases early.

Conclusion

This study proposes a new solution to precision agriculture by coupling AI-enabled autonomous agri- drones with intelligent smart soil sensor systems for real- time crop and soil health monitoring. The system is effective in bringing together aerial observation and ground sensing environmental data to deliver farmers with accurate, data-based information and automated decision- making Through the adoption of machine learning methods models like gradient boosting and LSTM networks, the system demonstrated accurate resource prediction as well as crop health prediction, simultaneously minimizing the usage of water and fertilizer. Efficient field coverage and rapid response to anomalies due to autonomy were achieved by the drone, resulting in enhanced crop yield as well as overall resource maximization. While the outcomes are encouraging, the system remains receptive to enhancements. Future work would be aimed at improving scalability with the use of swarm drone intelligence for mass operations. Integration of blockchain technology is to be pursued to improve transparency and traceability in the agricultural supply chain. The incorporation of edge AI onboard for real-time processing will reduce internet connectivity dependency, particularly in rural and remote areas. Enhancing energy efficiency with renewable-fueled sensor and drone modules, and designing more localized AI models for certain crops and geographies, will also be central agendas. Through these technologies, the suggested system can potentially be a central piece of future sustainable, autonomous, and intelligent agriculture. There are limitations, however. The model\'s accuracy is very vulnerable to lighting during flight activities and regional soil variability. Initial setup costs and regulatory limitation on drones may also hinder deployment in certain regions. To address these problems, the future work will involve edge-AI capabilities\' integration to enable offline analysis, expansion of the labeled dataset with local samples from areas for better generalization, and integration of explainable AI (XAI) to provide clear, human-understandable predictions. Broader implementation of this system can enable sustainable agricultural practices, particularly in data scarce or resource-constrained regions.

References

[1] H. Hammouch, M. A. El-Yacoubi, H. Qin, and H. Berbia, \"A Systematic Review and Meta-Analysis of Intelligent Irrigation Systems,\" IEEE Access, vol. 12, pp. 128285–128298, 2024. [2] B. Li, M. Thompson, T. Partridge, R. Xing, J. Cutler, B. Alhnaity, and Q. Meng, \"AI-Powered Digital Twin for Sustainable Agriculture and Greenhouse Gas Reduction,\" in Proc. 2024 IEEE 21st Int. Conf. on Smart Communities (HONET), pp. 55–61, 2024. [3] M. F. Almufareh, M. Humayun, Z. Ahmad, and A. Khan, \"An Intelligent LoRaWAN-Based IoT Device for checking and Control Solutions in Smart Farming Through Anomaly Detection Integrated With Unsupervised Machine Learning model,\" IEEE Access, vol. 13, pp. 119072–119090, 2025. [4] K. P. Reddy, V. Ebenezer, M. R. Thanka, S. Kirubakaran, E. B. Edwin, and P. Joy, \"Farm_era: Precision Farming with GIS, AI Pest Management, Smart Irrigation, Data Analytics, and Optimized Crop Planning,\" in Proc. 7th Int. Conf. on Inventive Computation Technologies (ICICT), pp. 1134–1141, 2024. [5] S. Nimmala, P. Manasa, M. Ramchander, M. A. Kiran, M. Mahendar, and B. Rambabu, \"A Recent Survey on AI Enabled Practices for Smart Agriculture,\" in Proc. 2024 Int. Conf. on Intelligent Systems for Cybersecurity (ISCS), pp. 1–6, 2024. [6] M. K. Bali and M. Singh, \"Farming in the Digital Age: AI-Infused Digital Twins for Agriculture,\" in Proc. 3rd Int. Conf. on Sentiment Analysis and Deep Learning (ICSADL), pp. 1–7, 2024. [7] M. Abdullah, S. Waheed, M. M. Hasan, and M. Morshed, \"Explainable AI-Driven Dew Point Forecasting With Attention- Based Temporal Convolutional Networks,\" IEEE Access, vol. 13, pp. 59923–59940, 2025. [8] B. Mosaad, R. Abdulla, and M. E. Rana, \"Recommendations for a Vertical Farming System Using Hydroponics, Machine Learning and IoT,\" in Proc. 2024 Int. Conf. on Green Technology and Sustainable Engineering (GTSE), pp. 1–7, 2024. [9] S. S. L. Chukkapalli, S. Mittal, M. Gupta, M. Abdelsalam, A. Joshi, R. Sandhu, and K. Joshi, \"Ontologies and Artificial Intelligence Systems for the Cooperative Smart Farming Ecosystem,\" IEEE Access, vol. 8, pp. 164045–164060, Sep. 2020. [10] R. J. Martin, R. Mittal, V. Malik, F. Jeribi, S. T. Siddiqui, M. A. Hossain, and S.L. Swapna, \"XAI-Powered Smart Agriculture Framework for Enhancing Food Productivity and Sustainability,\" IEEE Access, vol. 12, pp. 168412–168428, Nov. 2024. [11] A. Badshah, B. Y. Alkazemi, F. Din, K. Z. Zamli, and M. Haris, \"Crop Classification and Yield Prediction Using Robust Machine Learning Models for Agricultural Sustainability,\" IEEE Access, vol. 12, pp. 162799–162814, Oct. 2024. [12] T. Mahmud, N. Datta, R. Chakma, U. K. Das, M. T. Aziz, M. Islam, A. H. M. Salimullah, M. S. Hossain, and K. Andersson, \"An Approach for Crop Prediction in Agriculture: Integrating Genetic Algorithms and Machine Learning,\" IEEE Access, vol. 12, pp. 173583–173600, Oct. 2024. [13] R. Chithra, G. Thangavel, R. Vasantheeswaran, and A. P. Teijas, \"Enhancing Agricultural Efficiency and Sustainability Through Advanced IoT and AI-Driven Precision Farming Technologies,\" in 2024 International Conference on Electronics, Computing, Communication and Control Technology(ICECCCT),pp.1–6. [14] M. Bhasme, P. Ulhe, and S. Gokhe, \"AI in Agriculture: The Future of Robotic Farming and Automation,\" in 2024 2nd DMIHER International Conference on Artificial Intelligence in Healthcare, Education and Industry (IDICAIEI), pp. 1–6. [15] R. Jack, A. Banchit, H. Mohidin, M. Y. M. A. Khan, N. Dollah, K. F. Tamrin, and A. Narawi, \"Soil pH Mapping of Pineapple Crop: A Feasibility Study using Aerial Photo,\" 2019 IEEE, DOI: 10.1109, pp. 1–6 [16] P. Chattopadhyay, H. P. Patel, and V. Parmar, \"Internet of Things (IoT) in Smart Agriculture,\" Proceedings of the Third International Conference on Electronics and Sustainable Communication Systems (ICESC 2022), IEEE, 2022. [17] T. Bikov, G. Mihaylov, T. Iliev, and I. Stoyanov, \"Drone Surveillance in the Modern Agriculture,\" 2022 8th International Conference on Energy Efficiency and Agricultural Engineering (EE&AE), Ruse, Bulgaria, 2022. [18] R. K. Singh, R. Berkvens, and M. Weyn, \"AgriFusion: An Architecture for IoT and Emerging Technologies Based on a Precision Agriculture Survey,\" IEEE Access, vol. 9, pp. 136253– 136284, 2021. [19] S. Qazi, B. A. Khawaja, and Q. U. Farooq, \"IoT-Equipped and AI- Enabled Next Generation Smart Agriculture: A Critical Review, Current Challenges and Future Trends,\" IEEE Access, vol. 10, pp. 21219–21248, Mar. 2022. [20] R. Grati, N. Fattouch, and K. Boukadi, \"Ontologies for Smart Agriculture: A Path Toward Explainable AI—A Systematic Literature Review,\" IEEE Access, vol. 13, pp. 72883–72920, 2025. [21] A. Holzinger, I. Fister Jr., I. Fister Sr., H. P. Kaul, and S. Asseng, \"Human-Centered AI in Smart Farming: Toward Agriculture 5.0,\" IEEE Access, vol. 12, pp. 62199–62232, 2024.

Copyright

Copyright © 2025 Dr. Girish Kumar D, Miss. Allu Silpa. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.